Transfer learning for small datasets in TensorFlow - Model Pipeline Trace

Model Pipeline - Transfer learning for small datasets

This pipeline uses a pre-trained model to learn from a small dataset quickly and accurately. It reuses learned features from a large dataset and fine-tunes the model on new data.

Data Flow - 5 Stages

1Load small dataset

100 rows x 224 x 224 x 3→Load images and labels for 100 samples→100 rows x 224 x 224 x 3

Image of a cat with label 'cat'

↓

2Preprocessing

100 rows x 224 x 224 x 3→Resize and normalize pixel values to [0,1]→100 rows x 224 x 224 x 3

Normalized image pixel values between 0 and 1

↓

3Load pre-trained base model

224 x 224 x 3→Use MobileNetV2 without top layers, freeze weights→100 rows x 7 x 7 x 1280

Feature map tensor for each image

↓

4Add new classification head

100 rows x 7 x 7 x 1280→Global average pooling + dense layer for 5 classes→100 rows x 5

Logits for 5 classes per image

↓

5Train on small dataset

100 rows x 5→Fine-tune classification head with frozen base→Model weights updated for classification head

Model learns to classify new images

Training Trace - Epoch by Epoch

Loss
1.2 |*       
1.0 | *      
0.8 |  *     
0.6 |   *    
0.4 |    *   
    +---------
     1 2 3 4 5
     Epochs

Epoch	Loss ↓	Accuracy ↑	Observation
1	1.20	0.45	Model starts learning from small dataset
2	0.85	0.65	Loss decreases, accuracy improves
3	0.60	0.78	Model fine-tunes classification head well
4	0.50	0.82	Training stabilizes with good accuracy
5	0.45	0.85	Final epoch shows best performance

Prediction Trace - 5 Layers

Layer 1: Input image

Layer 2: Pre-trained base model (MobileNetV2)

Layer 3: Global average pooling

Layer 4: Dense classification layer

Layer 5: Softmax activation

Model Quiz - 3 Questions

Why do we freeze the base model weights during training?

A To increase the number of trainable parameters

B To keep learned features and avoid overfitting on small data

C To speed up training by skipping the classification head

D To prevent the model from making predictions

Key Insight

Transfer learning lets us use a big model trained on lots of data to quickly learn new tasks with small datasets. Freezing the base model keeps useful features, while training a small new head adapts to new classes efficiently.

Practice

1. What is the main reason to use transfer learning when working with small datasets in TensorFlow?

easy

A. To increase the size of the dataset automatically

B. To use a pretrained model's knowledge and avoid training from scratch

C. To reduce the number of layers in the model

D. To train all layers of a model from random weights

Solution

1. Step 1: Understand transfer learning purpose

   Transfer learning uses a pretrained model to leverage learned features from large datasets.

2. Step 2: Relate to small dataset challenges

   Small datasets lack enough data to train deep models well, so using pretrained knowledge helps.

3. Final Answer:

   To use a pretrained model's knowledge and avoid training from scratch -> Option B

4. Quick Check:

   Transfer learning = pretrained knowledge reuse [OK]

Hint: Transfer learning reuses pretrained features for small data [OK]

Common Mistakes:

• Thinking transfer learning increases dataset size
• Believing all layers must be retrained
• Assuming transfer learning reduces model layers

2. Which TensorFlow code snippet correctly freezes the base model layers for transfer learning?

easy

A. base_model.freeze_layers()

B. base_model.trainable = True

C. base_model.trainable = False

D. base_model.layers.freeze()

Solution

1. Step 1: Recall how to freeze layers in TensorFlow

   Setting trainable = False on the base model freezes all its layers.

2. Step 2: Check code options

   Only base_model.trainable = False is valid TensorFlow syntax to freeze layers.

3. Final Answer:

   base_model.trainable = False -> Option C

4. Quick Check:

   Freeze layers = trainable False [OK]

Hint: Set base_model.trainable = False to freeze layers [OK]

Common Mistakes:

• Using trainable = True instead of False
• Calling non-existent freeze_layers() method
• Trying to freeze layers with base_model.layers.freeze()

3. Given this code snippet, what will be the output shape of the model's predictions?

import tensorflow as tf
base_model = tf.keras.applications.MobileNetV2(input_shape=(128,128,3), include_top=False, pooling='avg')
base_model.trainable = False
model = tf.keras.Sequential([
  base_model,
  tf.keras.layers.Dense(10, activation='softmax')
])
model.build((None, 128, 128, 3))
predictions = model(tf.random.uniform((5, 128, 128, 3)))
print(predictions.shape)

medium

A. (5, 1)

B. (5, 128, 128, 3)

C. (5, 1000)

D. (5, 10)

Solution

1. Step 1: Analyze model architecture

   MobileNetV2 outputs pooled features (shape: (None, 1280)) due to pooling='avg'. Then Dense(10) outputs 10 classes.

2. Step 2: Check input batch size and output shape

   Input batch size is 5, so output shape is (5, 10) from Dense layer.

3. Final Answer:

   (5, 10) -> Option D

4. Quick Check:

   Batch 5, 10 classes output shape = (5, 10) [OK]

Hint: Output shape matches batch size and last Dense units [OK]

Common Mistakes:

• Confusing input shape with output shape
• Expecting output shape to match input image shape
• Assuming output has 1000 classes from original MobileNet

4. You wrote this code but get an error when training:

base_model = tf.keras.applications.ResNet50(weights='imagenet', include_top=False)
base_model.trainable = False
model = tf.keras.Sequential([
  base_model,
  tf.keras.layers.Dense(5, activation='softmax')
])
model.compile(optimizer='adam', loss='sparse_categorical_crossentropy')
model.fit(train_data, epochs=3)

What is the likely cause of the error?

medium

A. Missing input shape in base_model causing shape mismatch

B. trainable should be True to train the model

C. Loss function is incorrect for classification

D. Dense layer should have no activation

Solution

1. Step 1: Check base_model input shape requirement

   ResNet50 requires input_shape when include_top=False to build properly.

2. Step 2: Identify error cause

   Without input_shape, model layers don't connect correctly, causing training errors.

3. Final Answer:

   Missing input shape in base_model causing shape mismatch -> Option A

4. Quick Check:

   Input shape needed for base_model [OK]

Hint: Always specify input_shape when include_top=False [OK]

Common Mistakes:

• Not specifying input_shape for pretrained base model
• Thinking trainable must be True to avoid errors
• Changing loss function unnecessarily

5. You have a small dataset of 200 images for a 3-class classification task. Which transfer learning approach is best to get good accuracy quickly?

hard

A. Use a pretrained model, freeze base layers, add new Dense layers, train only new layers

B. Train a new model from scratch with random weights

C. Use a pretrained model and train all layers without freezing

D. Use a pretrained model without adding new layers and train only base layers

Solution

1. Step 1: Consider dataset size and training time

   Small dataset (200 images) benefits from freezing pretrained base to avoid overfitting and long training.

2. Step 2: Choose transfer learning strategy

   Freezing base layers and training new Dense layers adapts model to new task efficiently.

3. Final Answer:

   Use a pretrained model, freeze base layers, add new Dense layers, train only new layers -> Option A

4. Quick Check:

   Freeze base + train new layers = best for small data [OK]

Hint: Freeze base layers, train new layers for small datasets [OK]

Common Mistakes:

• Training all layers on small data causing overfitting
• Training from scratch wasting time and data
• Not adding new layers for the specific task